MCP低风险未认领

injectshield

Prompt-injection firewall for AI agents — scan untrusted text before LLM calls.

bch1212bch1212/injectshield更新于 2026年5月6日

server.json

{
  "$schema": "https://static.modelcontextprotocol.io/schemas/2025-09-29/server.schema.json",
  "name": "io.github.bch1212/injectshield",
  "description": "Prompt-injection firewall for AI agents — scan untrusted text before LLM calls.",
  "repository": {
    "url": "https://github.com/bch1212/injectshield",
    "source": "github",
    "subfolder": "packages/injectshield-mcp"
  },
  "version": "0.1.4",
  "websiteUrl": "https://injectshield.dev",
  "packages": [
    {
      "registryType": "npm",
      "registryBaseUrl": "https://registry.npmjs.org",
      "identifier": "@injectshield/mcp",
      "version": "0.1.4",
      "transport": {
        "type": "stdio"
      },
      "environmentVariables": [
        {
          "description": "API key from https://injectshield.dev (free tier: 10K req/mo).",
          "isRequired": true,
          "isSecret": true,
          "name": "INJECTSHIELD_API_KEY"
        },
        {
          "description": "Override the API base URL for self-hosted deployments.",
          "default": "https://api.injectshield.dev",
          "name": "INJECTSHIELD_API_BASE"
        }
      ]
    }
  ]
}

权限

声明检测
运行代码—node
安装—npm:@injectshield/mcp@0.1.4
安装时运行脚本—无
网络无无
需要的凭据INJECTSHIELD_API_KEYINJECTSHIELD_API_KEY
工作区外的路径—无
智能体工具—无

检查

低风险 · 没有发现需要提醒的地方。

未经人工审核 · 已做规则检查;模型审核尚未开启。

版本

  1. #10.1.4最新2026年10月7日